Detection of infrared dim and small targets based on saliency and grayscale morphological reconstruction

Based on saliency and grayscale morphological reconstruction, a new detection algorithm for infrared dim and small targets is proposed in this paper. The saliency of the original image to obtain the region of interest (ROI) is analyzed, then the spatial domain characteristic of dim and small targets is introduced into the marker image. Grayscale morphological reconstruction is based on the marker image and the mask image (the original image). Because saliency efficiently concentrates the gradient difference of the targets, detection probability is improved also with little false alarms. As the role of recognition, spatial domain characteristic of target reduces false alarm probability with the same detection probability, and background can be well estimated by grayscale morphological reconstruction, after subtraction, dim and small targets are detected. Experiments of real data prove the better detection performance, especially higher signal-to-noise ratio (SNR).